冠状动脉周围脂肪衰减指数(FAI)与粥样硬化斑块特征的相关性分析

Correlation Analysis of Pericoronary Fat Attenuation Index (FAI) and Atherosclerotic Plaque Characteristics

摘要: 目的：探讨冠状动脉周围脂肪衰减指数(Fat Attenuation Index, FAI)与冠状动脉动脉粥样硬化斑块特征之间的相关性。方法：回顾性分析99例存在冠状动脉粥样硬化患者的冠状动脉CT血管造影(Coronary Computed Tomography Angiography, CCTA)图像，其中221支冠状动脉存在斑块。使用人工智能软件自动测量存在斑块的冠状动脉血管近端40 mm周围脂肪FAI值，分析其与斑块相关特征参数包括斑块位置、斑块性质、斑块数量、斑块长度、血管狭窄程度的关系，使用线性回归分析FAI与斑块特征的相关性。结果：在冠状动脉粥样硬化患者中，对比不同血管分布、不同性质、不同狭窄程度、不同斑块长度的斑块发现，左回旋支FAI高于左前降支及右冠状动脉(P < 0.001)；钙化斑块所在血管FAI低于非钙化斑块(P < 0.05)及混合斑块(P < 0.005)；斑块呈弥漫性分布时血管FAI高于局限性分布(P < 0.05)及节段性分布(P < 0.005)；轻度狭窄血管FAI高于轻微狭窄(P < 0.001)及中度狭窄(P < 0.005)。多元线性回归分析表明，非钙化斑块、混合斑块，斑块位于左回旋支，斑块呈弥漫性分布均与冠周脂肪近端FAI值呈正相关(P < 0.05)。结论：FAI值与粥样硬化斑块所在位置、斑块类型以及斑块长度存在相关性，一定程度上对心血管病的诊断及预后有潜在的预测价值。

Abstract: Objective: To investigate the correlation between pericoronary fat attenuation index (FAI) and characteristics of coronary atherosclerotic plaques. Methods: Coronary Computed Tomography Angiography (CCTA) images of 99 patients with coronary atherosclerosis were retrospectively analyzed, of which 221 coronary arteries had plaques. Artificial intelligence software was used to automatically measure the FAI value of the fat 40 mm around the proximal end of the coronary artery with plaques, and its relationship with plaque-related characteristic parameters including plaque location, plaque nature, plaque number, plaque length, and degree of vascular stenosis was analyzed. Linear regression was used to analyze the correlation between FAI and plaque characteristics. Results: In patients with coronary atherosclerosis, the FAI of the left circumflex artery was higher than that of the left anterior descending artery and the right coronary artery (P < 0.001). The FAI of blood vessels in calcified plaques was lower than that in non-calcified plaques (P < 0.05) and mixed plaque (P < 0.005). The FAI of blood vessels in diffusely distributed plaques was higher than that in localized (P < 0.05) and segmental (P < 0.005) distributions. The FAI of vessels with mild stenosis was higher than that of vessels with mild stenosis (P < 0.001) and moderate stenosis (P < 0.005). Multivariate linear regression analysis showed that non-calcified plaques, mixed plaques, plaques located in the left circumflex artery, and diffusely distributed plaques were positively correlated with the proximal FAI value of pericoronary fat (P < 0.05). Conclusion: The FAI value is correlated with the location, type and length of atherosclerotic plaques, and has potential predictive value for the diagnosis and prognosis of cardiovascular disease to a certain extent.